ML Market Map — China A-Share Clusters for 2026-08-31

A daily unsupervised machine-learning read of the China A-Share market: 5,465 stocks grouped into 7 consensus clusters (KMeans + Gaussian-mixture + hierarchical, over robust-scaled PCA features) for 2026-08-31. Descriptive, not predictive — there is no buy or sell signal. Research, not investment advice.

Markets: US Equity · China A-Share · Japan Equity · Hong Kong Equity · Taiwan Equity · Korea Equity · UK Equity · India Equity

Today's clusters

• high margin · low volatility (1y) · low idio vol — 1,389 names; mostly Industrials; drivers: profit margin (+0.58), rv 252 (−0.55), residual vol 252 d (−0.53)
• falling (3m) · high downside vol · deep drawdown — 1,077 names; mostly Industrials; drivers: vol of vol 63 (+0.81), ret 63d (−0.58), downside vol 63 (+0.58)
• high leverage — 887 names; mostly Industrials; drivers: cfo growth (+1.29), leverage debt to mcap (+0.48), turnover to mcap (+0.32)
• strong 1y momentum · wide 1y upside range · favorable 1y edge — 626 names; mostly Technology; drivers: ret 252d (+2.56), pa mfe 52w (+1.86), ret 126d (+1.70)
• expensive (low E/P) · low margin · falling earnings — 561 names; mostly Industrials; drivers: val ep z (−2.55), profit margin (−2.15), ni growth (−1.44)
• high turnover · attention spent · wide upside excursions (21d) — 538 names; mostly Industrials; drivers: turnover to mcap (+2.31), pa sir recovered (+1.96), ret 20d (+1.30)
• high leverage · cheap (high B/P) · high cash yield — 387 names; mostly Industrials; drivers: leverage debt to mcap (+4.03), val bp z (+1.67), val cfop z (+1.54)

Machine-readable data (free, read-only JSON)

The full map, per-ticker cluster assignments with confidence and anomaly scores, and PCA structure are published as open JSON for automated and AI-analyst consumption:

• /api/ml/latest?market=cn — clusters, names, sector mix, representative tickers, top features
• /api/ml/tickers?market=cn — per-ticker cluster, confidence, anomaly score, day-over-day migration, tenure
• /api/ml/pca?market=cn — PCA components, explained variance, feature loadings

Descriptive market-structure research only. Unsupervised clustering finds structure, not direction; a tight cluster or an anomaly is a starting point for research, never a trade signal.